OWASP Agent Memory Guard
Open-source scanner for AI agent memory poisoning attacks (OWASP ASI06). Detects prompt injection payloads, memory manipulation patterns, and data exfiltration attempts in agent memory stores. Available as a Python package (pip install agent-memory-guard) and GitHub Action.
- from
- Microservices, Prompt Injection
- added
- 2026-10-10
- likes
- 0
Microservices › Capabilities > Security: “Runtime defense layer for AI agent memory poisoning (OWASP ASI06). Detects tampered memory entries, prompt injection in memory paths, and secret leakage. YAML policies, microsecond latency, zero external dependencies.”
Prompt Injection › Tools: “Open-source scanner for AI agent memory poisoning attacks (OWASP ASI06). Detects prompt injection payloads, memory manipulation patterns, and data exfiltration attempts in agent memory stores. Available as a Python package (pip install agent-memory-guard) and GitHub Action.”